Myocardial Viability (C0376314) in Patients (C0030705) with Chronic Coronary Artery Disease (C0010068) and Previous Myocardial Infarction (C0027051) : Comparison of Myocardial Contrast Echocardiography (C0013518) and Myocardial Perfusion (C0428857) Scintigraphy

Sarah Vernon, MD,Sanjiv Kaul, MD,Eric R. Powers, MD,Gustavo Camarano, MD,Lawrence W. Gimple, MD,Michael Ragosta, MD, Cardiovascular Division, University of Virginia, Charlottesville, Va.

013

American Heart Journal 11/01/97

Abstract

The aim (C0108800) of this study was to compare perfusion patterns (C0439712~C0449774) on myocardial contrast echocardiography (C0013518) with those on myocardial perfusion (C0428857) scintigraphy for the assessment of myocardial viability (C0376314) in patients (C0030705) with previous myocardial infarction (C0027051) .
Accordingly , perfusion scores (C0449820) with the two techniques (C0025664~C0449851) were compared in 91 ventricular regions (C0205147~C0446452) in 21 patients (C0030705) with previous ( > 6 weeks (C0439230~C0439506) old ) myocardial infarction (C0027051) .
Complete (C0205197) concordance between the two techniques (C0025664~C0449851) was found in 63 ( 69 % ) regions (C0205147~C0446452) ; 25 ( 27 % ) regions (C0205147~C0446452) were discordant by only 1 grade (C0441800~C0449383~C0456748) , and complete (C0205197) discordance ( 2 grades (C0441800~C0449383~C0456748) ) was found in only 3 ( 3 % ) regions (C0205147~C0446452) .
A kappa statistic (C0038215~C0220917) of 0.65 indicated good (C0004914~C0004916~C0004922~C0004923~C0005139~C0006030~C0053353~C0205170~C0332272~C0392746~C0439105~C0439127) concordance between the two techniques (C0025664~C0449851) .
Although the scores (C0449820) on both techniques (C0025664~C0449851) demonstrated a relation with the wall motion score (C0449820) , the correlation between the myocardial contrast echocardiography (C0013518) and wall motion scores (C0449820) was closer ( r (C0035253~C0175724~C0332153~C0332575~C0439117~C0439142) = -0.63 vs r (C0035253~C0175724~C0332153~C0332575~C0439117~C0439142) = -0.50 , p (C0016504~C0033452~C0036421~C0078414~C0439115~C0439140~C0449201~C0487890) = 0.05 ) .
It is concluded that myocardial contrast echocardiography (C0013518) provides similar information regarding myocardial viability (C0376314) as myocardial perfusion (C0428857) scintigraphy in patients (C0030705) with coronary artery disease (C0010068) and previous myocardial infarction (C0027051) .

Introduction

The extent (C0439792) of myocardial viability (C0376314) in patients (C0030705) with chronic coronary artery disease (C0010068) , previous myocardial infarction (C0027051) ( MI (C0026809~C0332288~C0347978) ) , and reduced left ventricular systolic function (C0031843~C0205245~C0422812) has both prognostic and therapeutic (C0039796~C0302350) significance .
Its assessment is therefore important in the clinical treatment (C0039798~C0087111) of such patients (C0030705) , especially when a revascularization procedure (C0025664~C0184661) is being considered .
Myocardial perfusion (C0428857) scintigraphy ( MPS ) is an established method (C0025663~C0025664) for the assessment of myocardial viability (C0376314) in the clinical setting (C0036849~C0324444) .
It is usually performed by measuring the uptake (C0243144~C0347980) of thallium- 201 ( which requires functional myocardial cell membranes (C0007603) ) or , less frequently , of Tc-labeled agents (C0450442) such as sestamibi ( which requires intact myocardial mitochondrial function (C0031843~C0205245~C0422812) ) .
Recently , myocardial contrast echocardiography (C0013518) ( MCE (C0286540) ) with intracoronary injection (C0021485) of microbubbles has been demonstrated to be useful for the assessment of myocardial viability (C0376314) in patients (C0030705) with both recent MI (C0026809~C0332288~C0347978) and those with poor left ventricular systolic function (C0031843~C0205245~C0422812) associated with chronic coronary artery disease (C0010068) .
The aim (C0108800) of this study was to compare perfusion patterns (C0439712~C0449774) on MCE (C0286540) with those on MPS for the assessment of myocardial viability (C0376314) in patients (C0030705) with chronic coronary artery disease (C0010068) and previous MI (C0026809~C0332288~C0347978) .

Methods (C0025663~C0025664)

Patient Population (C0032659)

The study was approved by the Human Investigation (C0430007) Committee at the University (C0041740) of Virginia (C0042753) , and all patients (C0030705) gave written informed consent .
Patients (C0030705) with previous MI (C0026809~C0332288~C0347978) ( > 6 weeks (C0439230~C0439506) old ) who had a corresponding wall motion abnormality (C0000768~C0000769) on left (C0205091~C0330089~C0450424) ventriculography underwent MCE (C0286540) at the time (C0040223~C0183941~C0332311~C0392761~C0449243) of diagnostic catheterization (C0007430) .
Twenty-one of these patients (C0030705) were also referred by their physicians (C0031831) for MPS within 4 weeks (C0439230~C0439506) of catheterization (C0007430) .
These 21 patients (C0030705) form the basis (C0004830~C0178499) of this report (C0178935~C0335038) , and their clinical (C0205210) characteristics are provided in Table (C0039224) 1 .

MCE (C0286540)

We have previously described the method (C0025663~C0025664) of performing MCE (C0286540) in the cardiac catheterization (C0018795~C0261588) laboratory .
In brief (C0205167~C0453896) , sonicated Renografin- (C0012005) 76 ( Squibb (C0331904) ) , which contains 500,000 + /-200,000 microbubbles of air (C0001861) with a mean diameter (C0332517) of 6 um (C0003818~C0180309~C0184514~C0230789~C0324145~C0439104~C0439126~C0439201) , was injected into the left main (C0205225) ( 1.5 ml (C0439242~C0439526) ) and right (C0035621~C0140586~C0205090) coronary ( 1.0 ml (C0439242~C0439526) ) arteries (C0003842) during simultaneously performed transthoracic (C0442386) two-dimensional echocardiography (C0013524) in multiple views (C0449911) ( mid-papillary muscle short-axis (C0522488) and apical four- and (C0205450) two-chamber views ) .
In patients (C0030705) with a totally occluded artery (C0003842) , contrast patterns (C0439712~C0449774) within the occluded bed (C0004914~C0004916~C0004922~C0004923~C0005139~C0006030~C0053353~C0205170~C0332272~C0392746~C0439105~C0439127) were noted during injection (C0021485) of the nonoccluded arteries (C0003842) , which result from collateral flow (C0330090) as described by us previously .

MPS

MPS was performed with planar tl imaging (C0150628) in 17 patients (C0030705) and Tc-sestamibi single photon emission (C0233929) computed tomography (C0040395) in the remaining 4 .
For 17 patients (C0030705) 3 mCi of tl (C0039738) was injected during exercise (C0015259~C0180776~C0452240) 5 minutes (C0205165~C0439232) before the initial (C0205265) images were obtained .
The delayed (C0205421) images were obtained 2 hours (C0439227) later .
Images were obtained in the anterior , 45-degree , and 70-degree left anterior oblique projections (C0442287) and were analyzed with a computer-assisted approach (C0449445) .
For four patients (C0030705) 8 mCi of Tc-sestamibi was injected at rest (C0035253~C0175724~C0332153~C0332575~C0439117~C0439142) , and images were acquired 1 hour (C0439227) later .
Exercise (C0015259~C0180776~C0452240) images were acquired later in the day (C0439228~C0439229~C0439505) 1 hour (C0439227) after injection (C0021485) of 25 to 30 mCi of Tc-sestamibi during peak exercise (C0015259~C0180776~C0452240) .
Data were processed with Ramp (C0034646~C0348053) and low-pass filters (C0180860~C0402833) and back-projection , after which tomographic images were created in the horizontal and vertical long-axis views (C0449911) .
Quantitative measurements (C0242485) were then performed in different myocardial regions (C0205147~C0446452) .
Only the delayed planar tl (C0039738) and the rest (C0035253~C0175724~C0332153~C0332575~C0439117~C0439142) Tc-sestamibi images were analyzed for this study .

Image Interpretation (C0459471)

The left ventricle (C0225897) was divided into five regions (C0205147~C0446452) for both forms (C0205431~C0348078~C0376315~C0426412) of imaging (C0150628) : apex (C0225409) and interventricular septum (C0225870) and lateral , inferior , and anterior walls (C0442070) .
Each region (C0205147~C0446452) was evaluated by two observers on MCE (C0286540) and MPS .
Similar scores (C0449820) were used for both methods (C0025663~C0025664) of imaging (C0150628) : 0 = no opacification (C0449584) on MCE (C0286540) and severe defect (C0243067~C0332451) ( < 50 % of maximal counts (C0076112~C0439157) ) on MPS ; 0.5 = partial or patchy opacification (C0449584) ( including that seen only in the epicardial (C0442016) rim ) on MCE (C0286540) and mild to moderate defect (C0243067~C0332451) ( 25 % to 50 % of maximal counts (C0076112~C0439157) ) on MPS ; and 1 = homogeneous opacification (C0449584) on MCE (C0286540) and normal uptake (C0243144~C0347980) on MPS .
The interobserver and intraobserver errors for MCE (C0286540) and MPS are small and have been previously reported .
Each region (C0205147~C0446452) was also graded for wall motion (C0026597) as follows : 1 = normal (C0205307~C0439166) , 2 = mild hypokinesia (C0086439) , 3 = severe hypokinesia (C0086439) , 4 = akinesia (C0085623) , and 5 = dyskinesia (C0013384) .
Our interobserver and intraobserver errors for wall motion score (C0449820) calculation are also small and have been previously reported .
The wall motion scores (C0449820) were assigned while the observers were blinded to the MCE (C0286540) and MPS data .

Statistical Methods (C0025663~C0025664)

All data were analyzed with RS (C0035253~C0175724~C0332153~C0332575~C0439117~C0439142) / 1 ( Bolt (C0336726) , Beranek , and Newman ) .
A weighted kappa statistic (C0038215~C0220917) was used to assess concordance between MCE (C0286540) and MPS .
The weight (C0005910~C0043100) favored fewer and penalized greater (C0205164~C0443228) differences in scores (C0449820) between the methods (C0025663~C0025664) .
Correlations between scores (C0449820) were performed with Spearman 's rank statistic (C0038215~C0220917) .

Results (C0332294)

Ninety-one of the 105 ventricular regions (C0205147~C0446452) were visualized by both techniques (C0025664~C0449851) .
Of the 14 regions (C0205147~C0446452) not visualized , 10 were because of inadequate MCE (C0286540) images where the interventricular septum (C0225870) was not seen in 4 , the lateral wall (C0205380) in 3 , the anterior wall (C0442070) in 2 , and the apex (C0225409) in 1 .
Four regions (C0205147~C0446452) were not well seen on MPS and included the interventricular septum (C0225870) in two and the lateral and anterior walls (C0442070) each in one .

Fig (C0349966) .
1 illustrates the concordance between the two tests (C0039593~C0039597~C0076174~C0154007~C0183885~C0392366~C0430738~C0496924) .
In the 91 regions (C0205147~C0446452) visualized by both techniques (C0025664~C0449851) , complete (C0205197) concordance was noted in 63 ( 69 % ) between the two methods (C0025663~C0025664) ; 25 regions (C0205147~C0446452) ( 27 % ) were discordant by only 1 grade (C0441800~C0449383~C0456748) , and complete (C0205197) discordance ( 2 grades (C0441800~C0449383~C0456748) ) was found in only three ( 3 % ) regions (C0205147~C0446452) .
A kappa statistic (C0038215~C0220917) of 0.65 indicated good (C0004914~C0004916~C0004922~C0004923~C0005139~C0006030~C0053353~C0205170~C0332272~C0392746~C0439105~C0439127) concordance between the two techniques (C0025664~C0449851) .

Fig (C0349966) .
2 is an example of concordance between the two techniques (C0025664~C0449851) and demonstrates an inferoapical defect (C0243067~C0332451) in a patient (C0030705) with previous inferior MI (C0026809~C0332288~C0347978) .
Although planar 201Tl shows (C0150769) overall low counts (C0076112~C0439157) in the inferior wall (C0205380) , the MCE (C0286540) image shows opacification (C0449584) limited only to the epicardial (C0442016) rim , with none seen in the rest (C0035253~C0175724~C0332153~C0332575~C0439117~C0439142) of the myocardium (C0027061) .
The inferior wall (C0205380) received a score (C0449820) of 0.5 by both techniques (C0025664~C0449851) .

In 16 regions (C0205147~C0446452) the MCE score (C0449820) was higher than the score (C0449820) on MPS , whereas in 12 regions (C0205147~C0446452) the converse occurred .
Most of the discordance occurred by 1 grade (C0441800~C0449383~C0456748) .
The three regions (C0205147~C0446452) with complete (C0205197) discordance ( > 1 grade (C0441800~C0449383~C0456748) ) included a lateral wall defect (C0243067~C0332451) seen on MPS but not on MCE (C0286540) and two apical defects (C0243067~C0332451) noted on MCE (C0286540) but not on MPS .
Thus discordance was noted in only 3 of the 21 patients (C0030705) .

An example of complete (C0205197) discordance between the two techniques (C0025664~C0449851) is depicted in Fig (C0349966) .
3 in a patient (C0030705) with a previous anteroapical MI (C0026809~C0332288~C0347978) .
A large perfusion defect (C0243067~C0332451) is seen in the apex (C0225409) on MCE (C0286540) , and no such defect (C0243067~C0332451) is seen in the apex (C0225409) on planar 201Tl imaging (C0150628) .
Although the 201Tl data are shown only in the 45-degree left anterior oblique projection (C0442287) , other projections (C0033363~C0348018) also did not show a perfusion defect (C0243067~C0332451) in the apex (C0225409) .

Although the scores (C0449820) on both MCE (C0286540) and MPS showed a correlation with wall motion score (C0449820) ( Fig (C0349966) .
4 ) , the correlation between the MCE (C0286540) and wall motion scores (C0449820) was closer ( p (C0016504~C0033452~C0036421~C0078414~C0439115~C0439140~C0449201~C0487890) = 0.63 vs p (C0016504~C0033452~C0036421~C0078414~C0439115~C0439140~C0449201~C0487890) = 0.50 , p (C0016504~C0033452~C0036421~C0078414~C0439115~C0439140~C0449201~C0487890) = 0.05 ) .

Discussion

Our study demonstrates that in patients (C0030705) with previous MI (C0026809~C0332288~C0347978) , MCE (C0286540) provides similar information regarding viability (C0376314) as MPS , a more established method (C0025663~C0025664) .
These findings (C0243095~C0263541~C0332285) potentially broaden the role (C0035820) of this technique (C0025664~C0449851) for the assessment of myocardial viability (C0376314) in the cardiac catheterization (C0018795~C0261588) laboratory .

Reasons (C0392360) for Concordance

The delayed 201tl image reflects the amount (C0205389) of 201Tl sequestered in the myocardium (C0027061) .
Because 201Tl enters the myocytes mostly by way of the Na/K pump (C0182537) , its presence (C0392148~C0392743) within the myocardium (C0027061) indicates that the cell membrane (C0007603) is intact .
Similarly , at rest (C0035253~C0175724~C0332153~C0332575~C0439117~C0439142) , Tc-sestamibi diffuses into the extravascular space (C0221874) and passively enters the myocyte before binding to the negatively charged mitochondrial membrane (C0025255) .
Consequently , its retention (C0035280~C0333117) within the myocyte is dependent on intact mitochondrial function (C0031843~C0205245~C0422812) .
The relative activity (C0439167~C0441655) of both tracers (C0442822) therefore provides an approximate (C0332232) estimate of the number (C0237753~C0449788) of myocytes that are viable within a region (C0205147~C0446452) .

In comparison , microbubbles used for MCE (C0286540) reside entirely within the vascular space (C0221874) .
They do not enter the extravascular space (C0221874) , nor are they extracted by myocytes .
They do not enter regions (C0205147~C0446452) where microvessels are absent such as in scar tissue (C0008767) .
Microbubbles are also not seen in regions (C0205147~C0446452) with other forms (C0205431~C0348078~C0376315~C0426412) of microvascular disruption (C0332453) , plugging , and obliteration (C0332465~C0391855) , which is frequently noted in areas (C0205146) with infarction (C0021308) .
The high degree (C0439489~C0449286) of concordance between MCE (C0286540) and MPS should therefore not be surprising and has also been found in studies with venous injections (C0021485) of microbubbles .
Mild (C0205080) differences in score (C0449820) ( 1 grade (C0441800~C0449383~C0456748) ) noted in our study can be explained on the different methods (C0025663~C0025664) of image representation ( tomographic vs planar ) .

Reasons (C0392360) for Discordance

Although complete (C0205197) discordance between the two techniques (C0025664~C0449851) was exceptional , it could be most dramatic as in the example shown in Fig (C0349966) .
3 .
There are several possible reasons (C0392360) for this discordance .
Although no clinically documented events (C0441471) occurred between cardiac catheterization (C0018795~C0261588) and MPS , there is a small chance (C0237506) that a new MI (C0026809~C0332288~C0347978) could have occurred .
The likelihood of an artifact (C0085089) on MCE (C0286540) is small , particularly if it encompasses the entire apex (C0225409) .
It is possible that the microvasculature (C0243079) within the apex (C0225409) was so sparse as not to be detected on MCE (C0286540) .
The low level (C0441889~C0456079) of myocardial perfusion (C0428857) may , however , still have allowed accumulation of 201Tl over several hours (C0439227) within viable myocardium (C0027061) .
Thus unless very low (C0442811) levels of flow (C0330090) are detectable on MCE (C0286540) , an extractable tracer (C0442822) may have an advantage (C0308269) in terms of assessing myocardial viability (C0376314) .
It is likely , however , that more sensitive methods (C0025663~C0025664) of detecting the presence (C0392148~C0392743) of microbubbles in tissue (C0040300) such as intermittent harmonic imaging (C0150628) will make it possible to measure low levels (C0441889~C0456079) of flow (C0330090) on MCE (C0286540) .

Microbubbles are destroyed on ultrasound exposure (C0274281~C0332157~C0391838) .
During the process of destruction (C0185043~C0240548) , they produce an " acoustic noise (C0028263) , " which contains many frequencies (C0042023~C0376249~C0439603) including the frequency (C0042023~C0376249~C0439603) to which they were exposed ( fundamental frequency (C0042023~C0376249~C0439603) ) .
Selective acquisition of the nonfundamental frequencies (C0042023~C0376249~C0439603) ( including harmonics of the fundamental frequency (C0042023~C0376249~C0439603) ) results in increased signal-to-noise ratio (C0456603) and better imaging (C0150628) of the microbubbles .
The signal-to-noise ratio (C0456603) is further increased when microbubble destruction (C0185043~C0240548) is minimized by obtaining images intermittently rather than continuously .
Experimental data from our laboratory (C0022877) indicate that this method (C0025663~C0025664) can be used to measure very low (C0442811) levels of flow (C0330090) .

Advantages (C0308269) of Contrast Echocardiography (C0013518)

In the cardiac catheterization (C0018795~C0261588) laboratory MCE offers a practical advantage (C0308269) over other techniques (C0025664~C0449851) in that it can provide more immediate viability (C0376314) assessment , which can then be used to guide clinical decision making (C0150516) and obviate the need (C0027552~C0028158~C0332125~C0347979~C0439138~C0439225~C0441922) for a noninvasive viability test (C0039593~C0039597~C0076174~C0154007~C0183885~C0392366~C0430738~C0496924) done on another day (C0439228~C0439229~C0439505) , which can increase hospital (C0019994) stay and cost (C0010186~C0220812) .

Because of its superior (C0205105) spatial resolution , MCE (C0286540) can define the transmural location (C0450429) of viable tissue (C0040300) , which can have prognostic implications .
For instance , in Fig (C0349966) .
2 both techniques (C0025664~C0449851) demonstrated partial viability (C0376314) .
However , on MCE (C0286540) it is clear that opacification (C0449584) is limited only in the epicardial (C0442016) rim .
Because wall thickening (C0205400~C0332527) is dependent on endocardial viability (C0376314) , this patient (C0030705) is unlikely to demonstrate improvement in regional function (C0031843~C0205245~C0422812) after revascularization .
The 201Tl image does not provide this potentially important information because it cannot distinguish between reduced perfusion (C0031001) caused by low flow (C0330090) from that caused by partial MI (C0026809~C0332288~C0347978) .

Limitations (C0449295) of the Study

Except for four patients (C0030705) , the MPS data were not tomographic .
Data registration between the two techniques (C0025664~C0449851) was therefore not optimal .
It is for this reason (C0392360) that we reduced the data to regions (C0205147~C0446452) of the left ventricular myocardium (C0027061) instead of comparing segments (C0442060) to each other (C0205394~C0220886~C0237094~C0237098~C0237111~C0237120~C0237127~C0237133~C0237142~C0237155~C0237161~C0237162~C0237169~C0237173~C0237178~C0237185~C0237190~C0237195~C0237206~C0237212~C0237221~C0237228~C0237237~C0237242~C0237243~C0237247~C0237255~C0237262~C0237266~C0237276~C0237283~C0237285~C0237289~C0237300~C0237305~C0237318~C0237324~C0237330~C0237338~C0237343~C0237344~C0237353~C0237357~C0237361~C0237366~C0237369~C0237373~C0237380~C0237387~C0237394~C0237395~C0449210) .
This approach (C0449445) , however , did not detract from the way we would have read each study clinically .
Despite the use (C0042153~C0439224~C0449865) of two different isotopes (C0022262) , no differences were seen in the results (C0332294) , although the number (C0237753~C0449788) of patients (C0030705) studied was small .

We did not perform a revascularization procedure (C0025664~C0184661) to determine which method (C0025663~C0025664) predicts recovery (C0237820) in function (C0031843~C0205245~C0422812) .
Revascularization was not clinically indicated in many of these patients (C0030705) , and follow-up studies were not performed in those who received such a procedure (C0025664~C0184661) .
We and others (C0205394~C0220886~C0237094~C0237098~C0237111~C0237120~C0237127~C0237133~C0237142~C0237155~C0237161~C0237162~C0237169~C0237173~C0237178~C0237185~C0237190~C0237195~C0237206~C0237212~C0237221~C0237228~C0237237~C0237242~C0237243~C0237247~C0237255~C0237262~C0237266~C0237276~C0237283~C0237285~C0237289~C0237300~C0237305~C0237318~C0237324~C0237330~C0237338~C0237343~C0237344~C0237353~C0237357~C0237361~C0237366~C0237369~C0237373~C0237380~C0237387~C0237394~C0237395~C0449210) have demonstrated the predictive value (C0042295~C0401806) of both techniques (C0025664~C0449851) for improvement of regional function (C0031843~C0205245~C0422812) after revascularization in similar kinds of patients (C0030705) .
This is the first (C0205435) study , however , that compares the results (C0332294) of the two techniques (C0025664~C0449851) in the same patients (C0030705) .

This study was performed in the cardiac catheterization (C0018795~C0261588) laboratory with selective coronary injections (C0021485) .
Recent (C0332185) experimental data indicate that similar results (C0332294) can be obtained with aortic root (C0489673) injections when intermittent harmonic imaging (C0150628) is used , which will make the technique (C0025664~C0449851) easier and simpler to perform .

Conclusions

MCE (C0286540) provides similar information as MPS for the putative assessment of myocardial viability (C0376314) in patients (C0030705) with coronary artery disease (C0010068) and old MI (C0026809~C0332288~C0347978) .
We and others (C0205394~C0220886~C0237094~C0237098~C0237111~C0237120~C0237127~C0237133~C0237142~C0237155~C0237161~C0237162~C0237169~C0237173~C0237178~C0237185~C0237190~C0237195~C0237206~C0237212~C0237221~C0237228~C0237237~C0237242~C0237243~C0237247~C0237255~C0237262~C0237266~C0237276~C0237283~C0237285~C0237289~C0237300~C0237305~C0237318~C0237324~C0237330~C0237338~C0237343~C0237344~C0237353~C0237357~C0237361~C0237366~C0237369~C0237373~C0237380~C0237387~C0237394~C0237395~C0449210) have previously demonstrated similar results (C0332294) in patients (C0030705) with recent MI (C0026809~C0332288~C0347978) .
These findings (C0243095~C0263541~C0332285) potentially broaden the role (C0035820) of this technique (C0025664~C0449851) for the assessment of myocardial viability (C0376314) in the cardiac catheterization (C0018795~C0261588) laboratory .

Presented in part (C0332285~C0449719) at the Seventh Annual (C0332181) Scientific Session of the American Society (C0037455) of Echocardiography (C0013516) , Chicago (C0008044) , Ill (C0105122~C0205169~C0332271) .
, June 10-12 , 1996 .

Reprint requests : Michael Ragosta , MD , Cardiovascular Division (C0441531) , Box (C0006080~C0179400) 158 , Medical Center (C0205099) , University (C0041740) of Virginia (C0042753) , Charlottesville , VA 22908 .

Table (C0039224) 1 - Clinical (C0205210) characteristics of the patient population (C0032659)